Search

Background & Aim Early clinical trials have demonstrated the safety of chimeric antigen receptor (CAR) T cells targeting glioblastoma (GBM), however, their efficacy remains limited by multiple obstacles including the highly immunosuppressive tumor microenvironment. These adoptively transferred CAR T cells are susceptible to inhibition via the engagement of inhibitory receptors on their surface including PD1, CTLA4, LAG3 and BTLA. The recruitment of Src homology region 2-containing protein tyrosine phosphatase 2 (SHP2) to the immune synapse upon engagement of these co-inhibitory receptors may represent a common mechanism of T cell inhibition. Upon its recruitment and activation, SHP2 dephosphorylates key stimulatory signaling molecules, thus limiting T cell activation. We hypothesized that disrupting SHP2 expression will simultaneously offset multiple co-inhibitory pathways, thereby improving the anti-tumor activity of CAR T cells. Methods, Results & Conclusion CRISPR/Cas9 technology was used to knockout (KO) SHP2 from human T cells. Retroviral vector transduction was used to express a clinically-utilized second generation CAR (with a CD28 signaling endodomain) targeting HER2. The phenotype of wild-type and edited CAR T cells was investigated using mass cytometry and confirmed with flow cytometry. Their function was tested in vitro using xcelligence (an electrical impedance-based long-term cytotoxicity assay) and in vivo in an orthotopic xenograft mouse model of GBM. Efficient SHP2 KO in human T cells was verified using western blotting. The Inference of CRISPR Efficiency (ICE) Assay confirmed efficient editing of the PTPN11 gene encoding SHP2. An anti-HER2 CAR was efficiently expressed in the SHP2KO T cells. SHP2 deletion did not significantly affect T cell expansion, proliferation or phenotypic profile. However, SHP2KO CAR T cells eliminated LN229 cells (a GBM cell line) in vitro more efficiently compared to wild-type CAR T cells. In vivo, SHP2KO CAR T cells showed better control of engrafted LN229 tumor cells and improved the survival of treated mice in comparison to wild-type CAR T cells. In conclusion, these results suggest that SHP2 may limit the function of CAR T cells and that SHP2 deletion may improve their anti-GBM activity.

Epstein-Barr virus (EBV)-specific cytotoxic T lymphocytes (CTLs) can be modified to function as heterologous tumor directed effector cells that survive longer in vivo than tumor directed T cells without virus specificity, due to chronic stimulation by viral antigens expressed during persistent infection in seropositive individuals. We evaluated the nonviral piggyBac (PB) transposon system as a platform for modifying EBV-CTLs to express a functional human epidermal growth factor receptor 2-specific chimeric antigen receptor (HER2-CAR) thereby directing virus-specific, gene modified CTLs towards HER2-positive cancer cells. Peripheral blood mononuclear cells (PBMCs) were nucleofected with transposons encoding a HER2-CAR and a truncated CD19 molecule for selection followed by specific activation and expansion of EBV-CTLs. HER2-CAR was expressed in ~40% of T cells after CD19 selection with retention of immunophenotype, polyclonality, and function. HER2-CAR-modified EBV-CTLs (HER2-CTLs) killed HER2-positive brain tumor cell lines in vitro, exhibited transient and reversible increases in HER2-CAR expression following antigen-specific stimulation, and stably expressed HER2-CAR beyond 120 days. Adoptive transfer of PB-modified HER2-CTLs resulted in tumor regression in a murine xenograft model. Our results demonstrate that PB can be used to redirect virus-specific CTLs to tumor targets, which should prolong tumor-specific T cell survival in vivo producing more efficacious immunotherapy.

Human epidermal growth factor receptor 2 (HER2) is expressed by the majority of human osteosarcomas and is a risk factor for poor outcome. Unlike breast cancer, osteosarcoma cells express HER2 at too low, a level for patients to benefit from HER2 monoclonal antibodies. We reasoned that this limitation might be overcome by genetically modifying T cells with HER2-specific chimeric antigen receptors (CARs), because even a low frequency of receptor engagement could be sufficient to induce effector cell killing of the tumor. HER2-specific T cells were generated by retroviral transduction with a HER2-specific CAR containing a CD28.zeta signaling domain. HER2-specific T cells recognized HER2-positive osteosarcoma cells as judged by their ability to proliferate, produce immunostimulatory T helper 1 cytokines, and kill HER2-positive osteosarcoma cell lines in vitro. The adoptive transfer of HER2-specific T cells caused regression of established osteosarcoma xenografts in locoregional as well as metastatic mouse models. In contrast, delivery of nontransduced (NT) T cells did not change the tumor growth pattern. Genetic modification of T cells with CARs specific for target antigens, expressed at too low a level to be effectively recognized by monoclonal antibodies, may allow immunotherapy to be more broadly applicable for human cancer therapy.

BACKGROUND: We demonstrated that GBM patients’ T cells expressing HER2-specific chimeric antigen receptors (CAR) killed autologous HER2 + GBM cells and induced regression of autologous xenografts; however, tumors recurred in a number of treated animals, likely due to limited T-cell persistence and/or tumor antigen loss variants. Since the majority of GBMs express CMVpp65, we expressed HER2.CARs in CMV-specific cytotoxic T cells (CTLs) to enhance their persistence following native T-cell receptor (alphabetaTCR) engagement with CMV-antigens and improve their antitumor activity by simultaneously targeting CMV and HER2. METHODS: To study the feasibility of generating CMV CTLs from GBM patients, we determined the precursor frequency of CMV-specific T cells in 25 patients using IFN-gamma-ELIspot assays. To generate CMV CTL, blood mononuclear cells were transduced with an adenovirus vector, encoding the immunodominant CMV-pp65 antigen (Ad5f35pp65) and then stimulated with autologous EBV-transformed Ad5f35pp65-transduced irradiated lymphoblastic cell lines (EBV-LCL). CTLs were then transduced with a retroviral vector encoding a HER2.CAR. We detected the expression of the CAR using flow cytometry. To concomitantly identify the CMV-specific T cell population, we used pentamers for CMVpp65 and performed Cr-release assays to test their functionality. RESULTS: CMV-specific T cells were readily detectable in the blood of GBM patients, as judged by IFN-gamma secretion after stimulation with CMVpp65-PepMixes. We were able to generate CTLs that are specific for CMV through their endogenous alphabetaTCR and specific for HER2 through CAR. Approximately 50% of CMVpp65.CTLs expressed HER2.CARs on the cell surface. CMV.HER2.CAR CTLs recognized and killed the HER2+GBM cells as well as autologous T cells pulsed with CMVpp65-PepMixes in Cr-release assays. Neither the HER2 cell lines nor autologous T cells were killed. CONCLUSION: Our data demonstrate the feasibility of generating CAR-expressing CMVpp65– specific CTLs that are functional against HER2 and CMV expressed in GBM.